CODEWITHSUNDEEP
rdplyr
KillerSnail
KillerSnail

Reputation: 3591

pivot column into row titles in R?

I have a dataframe

division | category
  A      |    tools
  A      |    work
  B      |    tools
  B      |    books

both columns are factor variables. how do I pivot the category columns so that I get:

division | tools | books | work
   A         1               1
   B         1       1

I tried using dplyr

wide <- df %>%
    spread(division, category)

but I still get the same number of rows it hasn't collapsed them down to open per division?

Upvotes: 2

Views: 2088

Answers (1)

akrun
akrun

Reputation: 887881

One option would be to create another column of 1 with transform, and use that column as the value.var in dcast

library(reshape2)
dcast(transform(df, ind=1), division~category, value.var='ind')

Or as @MichaelChirico mentioned, we can specify fun.aggregate as length

dcast(df, division~category, length)

Or if we are using dplyr/tidyr, mutate the dataset to create the new column and then spread from 'long' to 'wide'. spread always needs the column that contains the value to put into the output data.frame. This was not present in the example of the OP, and adding it solves the issue.

library(dplyr)
library(tidyr)
df %>% 
   mutate(ind=1) %>% 
   spread(category,ind)

Upvotes: 5

Related Questions